ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-04 05:31:24 +00:00
Code Issues Packages Projects Releases Wiki Activity

[CK_TILE] Stream-K Tile Engine Test Config File Generation (#3662)

Browse Source 
* Stream-K smoke test config file generation

This change converts the stream-k smoke tests to use tile engine. Since
the m, n, and k values dependent on the CU count of a device, the
configs are generated during the Configuration Phase.

* Compute GEMM reference on GPU

* Remove redundant Stream-K tests

Removing redundant tests that are now run via tile engine.

* Fix relative and absolute tolerance calculation

This change updates the Stream-K tile engine interface to ensure that
num_wgs_per_tile is propaged and passed into the compare_results
function to calculate the rel and abs tolerance. Before, split-k was
used, which is incorrect for Stream-K since the split-k value is
always 1.

* Cleanup imports, types, and other misc items

This commit makes the following changes:
- Uses Typing module for nested type hints
- Uses quotes around cu_count_arg argument in generate_configs.cmake in
  if statements
- Adds explicit include for tuple in test_gemm_streamk_simple.cpp
- Adds a type for the tiles argument in argparser to check argument
  validity

* Use CU count as return value for better parsing

* Add reduction tests for bf16, fp8, and bf8
This commit is contained in:
Emily Martins
2026-02-03 09:12:15 -07:00
committed by GitHub
parent 3f04d27b68
commit 8cbd09c84a
22 changed files with 522 additions and 406 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
test/ck_tile/gemm_streamk/CMakeLists.txt
View File

@@ -23,19 +23,6 @@ if(GPU_TARGETS MATCHES "gfx90a|gfx942|gfx950")
#TODO: support all arches
#TODO: current c-shuffle only supports C layout as R
add_gtest_executable(test_ck_tile_streamk_tile_partitioner test_streamk_tile_partitioner.cpp)
add_gtest_executable(test_ck_tile_streamk_reduction
${CMAKE_CURRENT_SOURCE_DIR}/smoke_tests/test_gemm_streamk_fp16_reduction.cpp
test_gemm_streamk_util.cpp)
add_gtest_executable(test_ck_tile_streamk_smoke
${CMAKE_CURRENT_SOURCE_DIR}/smoke_tests/test_gemm_streamk_fp16_persistent.cpp
${CMAKE_CURRENT_SOURCE_DIR}/smoke_tests/test_gemm_streamk_bf16_persistent.cpp
${CMAKE_CURRENT_SOURCE_DIR}/smoke_tests/test_gemm_streamk_fp8_persistent.cpp
${CMAKE_CURRENT_SOURCE_DIR}/smoke_tests/test_gemm_streamk_bf8_persistent.cpp
${CMAKE_CURRENT_SOURCE_DIR}/smoke_tests/test_gemm_streamk_fp16_nonpersistent.cpp
${CMAKE_CURRENT_SOURCE_DIR}/smoke_tests/test_gemm_streamk_bf16_nonpersistent.cpp
${CMAKE_CURRENT_SOURCE_DIR}/smoke_tests/test_gemm_streamk_fp8_nonpersistent.cpp
${CMAKE_CURRENT_SOURCE_DIR}/smoke_tests/test_gemm_streamk_bf8_nonpersistent.cpp
test_gemm_streamk_util.cpp)
add_gtest_executable(test_ck_tile_streamk_extended
${CMAKE_CURRENT_SOURCE_DIR}/extended_tests/test_gemm_streamk_fp16_persistent.cpp
${CMAKE_CURRENT_SOURCE_DIR}/extended_tests/test_gemm_streamk_bf16_persistent.cpp
@@ -46,7 +33,6 @@ if(GPU_TARGETS MATCHES "gfx90a|gfx942|gfx950")
${CMAKE_CURRENT_SOURCE_DIR}/extended_tests/test_gemm_streamk_fp8_nonpersistent.cpp
${CMAKE_CURRENT_SOURCE_DIR}/extended_tests/test_gemm_streamk_bf8_nonpersistent.cpp
test_gemm_streamk_util.cpp)
target_compile_options(test_ck_tile_streamk_smoke PRIVATE ${EXAMPLE_GEMM_COMPILE_OPTIONS})
target_compile_options(test_ck_tile_streamk_extended PRIVATE ${EXAMPLE_GEMM_COMPILE_OPTIONS})
else()
message(DEBUG "Skipping test_ck_tile_streamk unit tests for current target")